Averaging Weights Leads to Wider Optima and Better Generalization

Pavel Izmailov, Dmitrii Podoprikhin, Timur Garipov, Dmitry P. Vetrov, Andrew Gordon Wilson. Averaging Weights Leads to Wider Optima and Better Generalization. In Amir Globerson, Ricardo Silva, editors, Proceedings of the Thirty-Fourth Conference on Uncertainty in Artificial Intelligence, UAI 2018, Monterey, California, USA, August 6-10, 2018. pages 876-885, AUAI Press, 2018. [doi]

Authors

Pavel Izmailov

This author has not been identified. Look up 'Pavel Izmailov' in Google

Dmitrii Podoprikhin

This author has not been identified. Look up 'Dmitrii Podoprikhin' in Google

Timur Garipov

This author has not been identified. Look up 'Timur Garipov' in Google

Dmitry P. Vetrov

This author has not been identified. Look up 'Dmitry P. Vetrov' in Google

Andrew Gordon Wilson

This author has not been identified. Look up 'Andrew Gordon Wilson' in Google